The study, led by Fan-Chi Hsiao of Taipei Medical University and Ming Chuan University, together with colleagues at National Yang Ming Chiao Tung University, set out to test a deceptively simple hypothesis. Gender-role attitudes, the beliefs people hold about what kinds of work and family responsibilities are appropriate for men and women, may shape how employed parents interpret the demands of their dual roles. The researchers proposed that parents with more traditional attitudes would experience poorer subsequent sleep, and that this effect would be stronger for mothers than for fathers. The reasoning draws on a substantial body of work suggesting that traditional gender norms tend to place heavier domestic and caregiving burdens on women, even when both partners work full time, and that such unequal burdens are a recognized correlate of disturbed sleep.

To investigate, the team recruited 523 full-time employed parents in Taiwan, each with at least one child aged twelve or younger. Participants were assessed at baseline and then again approximately three and six months later, a three-wave design that allows researchers to separate earlier attitudes from later sleep outcomes in a way that single-snapshot surveys cannot. Sleep quality was measured with the Chinese Pittsburgh Sleep Quality Index, a validated instrument that captures multiple dimensions of sleep, including duration, disturbances, latency, and subjective quality. Gender-role attitudes were assessed with items drawn from established measures, including instruments connected to the International Social Survey Programme and the Taiwan Social Change Survey, supplemented by scales addressing work-family conflict and life-role salience.

The primary analysis used linear mixed models, a statistical framework well suited to repeated-measures data because it can model change within individuals while accounting for variation between them. In the model that included 478 participants with at least one follow-up assessment and complete covariate data, the researchers predicted follow-up sleep quality scores from baseline gender-role attitudes, parent gender, and the interaction between the two, while adjusting for baseline sleep and demographic characteristics. This adjustment for baseline sleep is methodologically crucial: it asks whether attitudes predict a change in sleep over time, rather than simply reflecting the fact that people who already sleep poorly may hold particular views.

The headline result was an interaction. After accounting for sleep quality at study entry, the association between gender-role attitudes and later sleep differed between mothers and fathers in the direction the researchers had hypothesized, with an interaction coefficient of 0.126 and a p-value of .028. In practical terms, the estimated relationships ran in opposite directions for the two groups: the pattern of associations suggested that attitudes related to sleep differently depending on whether the parent was a mother or a father. Notably, however, neither of the two group-specific associations reached statistical significance on its own in the primary model. The evidence, in other words, points to a difference between the sexes rather than to two independently confirmed effects.

The authors are careful, and rightly so, about how far this finding can be pushed. When the analysis was repeated without adjusting for baseline sleep, or when it was averaged across all three assessment waves, the gender difference shrank to nearly zero. The hypothesized pattern appeared only under one specific longitudinal approach, and the researchers describe it as exploratory. This kind of inconsistency across analytic strategies is a familiar feature of behavioral research, and the study’s transparency about it is a model of how such results should be reported. A single significant interaction that does not replicate across reasonable alternative specifications should be treated as a lead worth following, not as an established fact.

There is a second, more technical caveat that shapes how the findings should be read. Before running the primary models, the team conducted ordered-categorical analyses to examine the measurement properties of the gender-role attitude items and whether they functioned comparably across mothers and fathers. They found that the items were not fully comparable between the two groups, a problem known in psychometrics as a lack of measurement invariance. When a questionnaire item means something different to one group than to another, differences in scores between those groups can reflect differences in interpretation rather than differences in the underlying attitude. The authors acknowledge that this limited comparability leaves the substantive meaning of their gender difference uncertain, a candor that strengthens rather than weakens the study’s contribution.

Why should attitudes about gender be connected to sleep at all? The most plausible mechanisms run through the architecture of daily life. A parent who believes that childcare and housework are primarily a woman’s responsibility may organize evenings, nights, and early mornings differently from one who sees those duties as shared. For mothers, traditional attitudes could coincide with accepting a larger share of nighttime caregiving, interrupted sleep, and the mental load of household coordination. For fathers, the same attitudes might align with a different configuration of work and family demands, potentially one that protects sleep or, alternatively, one tied to provider stress. The study did not directly measure how caregiving and household labor were divided, so these pathways remain hypotheses rather than demonstrated mechanisms.

The Taiwan setting gives the work particular interest. Taiwanese society has undergone rapid change in gender norms over recent decades, with high female labor-force participation coexisting alongside persistently traditional expectations about family roles. That tension makes the island an informative natural laboratory for questions about how evolving attitudes interact with entrenched domestic arrangements. The study was supported by the National Science and Technology Council, Taiwan, and received ethical approval from the Research Ethics Committee of National Chengchi University, with all participants providing informed consent. The authors also note that generative AI tools were used only for language refinement and code checking, with all analyses independently designed, conducted, and verified by the researchers.

What the study ultimately offers is a well-framed question rather than a settled answer. Its conclusion points toward a promising next step: future research should directly measure how daily caregiving and household work are actually shared between partners, because the same attitudes may relate to sleep very differently depending on the everyday circumstances in which they are held. For now, the finding that gender-role attitudes and later sleep may diverge between mothers and fathers, observed after accounting for where each parent started, invites both researchers and clinicians to think about sleep not merely as a product of hours and schedules, but as something woven into the beliefs and bargains that structure family life. If subsequent studies confirm and clarify this pattern, the humble questionnaire about gender roles could become an unexpected tool for understanding who in a household gets a good night’s rest, and why.

Subject of Research: The longitudinal relationship between gender-role attitudes and sleep quality among full-time employed mothers and fathers in Taiwan

Article Title: Gender-role attitudes and sleep among full-time employed mothers and fathers in Taiwan: a three-wave longitudinal study

Article References: Hsiao, F.-C., Lin, M.-T., Huang, S.-H., & Lin, H.-T. (2026). Gender-role attitudes and sleep among full-time employed mothers and fathers in Taiwan: a three-wave longitudinal study. BMC Psychology. https://doi.org/10.1186/s40359-026-05681-1

Image Credits: AI Generated

DOI: 10.1186/s40359-026-05681-1

Keywords: gender-role attitudes, sleep quality, working parents, longitudinal study, Taiwan, work-family conflict, Pittsburgh Sleep Quality Index, mothers, fathers, measurement invariance, gender and health, parental sleep